Vel Tech Rangarajan Dr. Sagunthala R&D Institute of Science and Technology

Categorical variants, or sets of genomic alterations constrained by shared properties, are pervasive across clinical, regulatory, and research domains in the biomedical ecosystem, yet their inconsistent and non-computable representation hinders data interoperability and clinical interpretation. We surveyed genomic knowledgebases spanning regulatory approvals and the biomedical literature and found that categorical variants underpin a substantial proportion of clinical genomics knowledge, but are largely described using incompatible bespoke models. To address this, we developed the GA4GH Categorical Variation Representation Specification (Cat-VRS), a constraint-based framework that provides a unified computable representation for both precise and intentionally broad categories across molecular and systemic variant domains. Cat-VRS enables harmonization of genomic knowledgebases, computable category-based search, and automated matching between assayed variants and categorical entities in clinical and research contexts. By providing a principled, extensible model for categorical variation, Cat-VRS enables computable reasoning over genomic variant categories and establishes a foundation for the standardized representation and exchange of genomic knowledge.

Social media websites are popular, and it is now possible to collect a significant amount of behavioral data in real time. This has made it possible to conduct research on mental health. This study proposes a machine learning framework to identify and predict mental health issues, such as emotion and cognition, related to social media with respect to depression, anxiety, and bipolar disorder. In this study, three social media datasets were used, each with more than 1.6 million posts: Sentiment140, Twitter Depression Dataset, and Facebook Sentiment. The data was tokenized and lemmatized, and stopwords were removed. We used BERT, a transformer model, to view the data using emotional traits. This was done using the sentimental polarity score and emotions such as grief, anger, joy, fear, and insignificance. We used Latent Dirichlet Allocation (LDA) and Linguistic Inquiry and Word Count (LIWC) to infer cognitive distortions and psychological indices. We used Recursive Feature Elimination (RFE) for feature selection and a Random Forest classifier to predict mental health across multiple classes. The proposed method is more effective than deep learning and machine learning techniques, achieving an accuracy of 95.0%, precision of 94.0%, recall of 93.7%, F1-score of 94.8%, and AUC-ROC of 98.9%. The sentiment polarity score improved the prediction by 28%, the emotional scores improved the prediction by 35%, the cognitive themes improved the prediction by 22%, and the LIWC features improved the prediction by 15%. The results indicate that the use of cognitive linguistics and emotional representation facilitates the identification of early symptoms of severe depression and bipolar disorder.

This work investigates the mechanical, impact, shear, and morphological properties of epoxy-based composites reinforced with 40 vol.% Pterospermumacerifolium stem fiber and Hordeumvulgare husk-derived silicon oxynitride. The addition of fibers increased the tensile strength from 53 MPa (neat epoxy) to 109 MPa and flexural strength from 68 MPa to 128 MPa, while the incorporation of silicon oxynitride as a nano-filler further enhanced the properties. Among the composites, EPS2 (3 vol.% silicon oxynitride) exhibited the highest mechanical performance, with tensile strength of 145 ± 4.4 MPa, flexural strength of 161 ± 5.1 MPa, and Izod impact strength of 4.6 ± 0.14 J, demonstrating optimal fiber-matrix interaction and reinforcement efficiency. In high-velocity impact testing, EPS2 absorbed 6.8 ± 0.24 J with an extended impact duration of 14.3 ± 0.6 ms, while total energy absorption reached 38.7 ± 1.3 J with a deflection of 3.8 ± 0.14 mm. EPS2 also displayed the highest V-Notch shear strength (21.4 ± 0.8 MPa) and rail shear strength (22.4 ± 0.9 MPa), confirming enhanced shear load-bearing capacity. SEM analysis revealed uniform filler dispersion and strong fiber-matrix adhesion, minimizing voids and enhancing mechanical stability, whereas excessive filler content in EPS3 led to agglomeration and reduced performance. Overall, EPS2 demonstrated the best balance of strength, toughness, and impact resistance, showing minimal perforation, negligible delamination, and superior energy dissipation, making it the most suitable composition for structural applications requiring high mechanical integrity.

The fast development of the applications of the Industrial Internet of Things (IIoT) requires the real-time object detection which has the capability to work effectively in energy-restrained edgefog conditions. Unlike the former YOLO-based and lightweight detectors and RL-based edgefog offloading mechanisms, the proposed RAEED framework combines lightweight detection and training-free adaptive inference offloading to optimize the accuracy, latency and energy consumption jointly in IIoT settings. The deep learning-based detectors are very accurate and capable of handling resource-intensive IIoT deployments, however, their systems are frequently computation- and powerintensive, which is not feasible in resource-constrained IIoT applications. In this effort, the present work suggests that a resource-conscious energy-efficient detector (RAEED) system is proposed which allocates inference resources between the edge and fog nodes dynamically projected by resource availability and network characteristics. The framework integrates a convolutional backbone which is lightweight together with adaptive offloading strategy to trade-off between detection accuracy, latency, and energy consumption. The applications to the COCO 2017 dataset under constrained deployment conditions indicate that RAEED can have 82.4% mAP, 83.1% precision and 81.7% recall, with a latency of 47.6 ms and with a consumption of 0.91 mJ/frame. These findings indicate that IIoT systems can greatly manage to find and identify objects with a much better energy and resource savings.

Artificial intelligence (AI)-powered technology integration in social fintech has transformative potential to advance social responsibility and support sustainable development. This research examines a Blockchain-based lending mechanism that integrates centralized exchanges (CEX) and decentralized exchanges (DEX) to facilitate seamless financial transactions and equitable resource allocation. AI-driven tools are utilized to enhance transparency, accuracy, and security, while smart contracts facilitate the efficient management and verification of loan distribution. The proposed system focuses on helping underserved communities, poor regions, and green businesses, promoting fair and sustainable finance in line with the Sustainable Development Goals (SDGs). The hybrid ecosystem combines the liquidity and regulatory compliance of centralized exchanges with the autonomy and reduced intermediary involvement of decentralized exchanges. AI enhances loan processing, reducing biases and inefficiencies. This framework with smart contracts is to provide scalable, auditable lending aligned with sustainable goals. Machine Learning (ML) algorithms verified loan eligibility with the borrower dataset. The performance of Random Forest algorithms is good due to their robustness and ensemble learning features. Then, Optuna enhanced model tuning, and SHapley Additive exPlanations (SHAP) identified key parameters. Finally, Smart contracts ensured secure, autonomous execution of green loans based on ML verification and sustainability criteria.

This investigation reports the fabrication and comprehensive functional evaluation of a newly engineered hybrid nanocomposite composed of gallium nitride (GaN) integrated with a poly(3-methylthiophene)-co-poly(indole) (P3MTH–PID) copolymer, synthesized via a straightforward chemical oxidative polymerization route. GaN nanoparticles were prepared using a supercritical ammonia method and subsequently combined with P3MTH–PID nanofibers to form a uniform hybrid structure. Structural and surface analyses confirmed robust π–π stacking interactions between the GaN scaffold and the P3MTH–PID copolymer matrix, which facilitated efficient charge delocalization and ensured homogeneous nanofiber distribution across the GaN interface. The hybrid material exhibited remarkable electrocatalytic sensitivity toward albendazole (ABZ), as demonstrated by differential pulse voltammetry (DPV), achieving a detection limit (DL) of 1.372 × 10⁻⁸ M µA⁻¹ and a quantification limit (QL) of 4.352 × 10⁻⁸ M µA⁻¹. Furthermore, the hybrid displayed excellent photocatalytic performance, enabling complete discoloration of methylene blue (MB) within 30 minutes under UV–visible irradiation, with a kinetic rate constant of 78.66 × 10⁻² min⁻¹. Beyond sensing and photocatalysis, the GaN–P3MTH–PID hybrid also showed significant promise in energy storage applications. In a three-electrode configuration, it delivered a high specific capacitance of 294.25 F g⁻¹ at 1 A g⁻¹, while a symmetric supercapacitor device (2 × 2 cm²) assembled from the same material exhibited comparable capacitance and excellent long-term cycling stability. Collectively, these findings highlight that the GaN–P3MTH–PID hybrid integrates superior electrochemical, photocatalytic, and sensing functionalities with strong stability and biocompatibility, making it a highly versatile candidate for sustainable and multifunctional energy storage and environmental applications.

Federated Learning (FL) enables accurate and secure Clinical Event Prediction (CEP) across distributed hospitals. However, the prevailing works overlooked the catastrophic forgetting during the global update. Therefore, a Meta Experience Polynomial Decay-based Replay (MEPDR)-centric continual update is proposed. Initially, the hospitals (local model) register and log into the blockchain. Then, to train the CEP model, data collection, pre-processing, and feature extraction are performed. Further, the Temporal-Causal Graph (TCG) is constructed. Afterward, the node matrix is created, and the CEP is done using Mean-Centering Normalization-based Graph Neural Network (MCN-GNN). The model's gradients are further preserved using the Homomorphic Robust Log Scaling-based Encryption (HRLSE). Next, the hospitals are authenticated using the Exponential Probing Digital Signature Algorithm (ExPrDSA). Thereafter, in the global model, the aggregation is performed using the Calinski-Harabasz Index with Zhonghua Distance-based K-Means Clustering (CHIZD-KMC), followed by global CEP. After that, during the global update, the MEPDR-based continual learning is carried out in each local model. Also, the transactions are stored in the blockchain to enhance traceability. Thus, the proposed system effectively predicted the clinical events with an accuracy of 98.97%, outperforming existing works.

In an era of digital disruption and hyper-personalized marketing, the integration of artificial intelligence (AI) into market segmentation and consumer behavior modeling has revolutionized strategic branding. This chapter explores how machine learning algorithms, particularly unsupervised clustering, neural networks, and predictive analytics, can uncover hidden consumer patterns and dynamically segment markets with unparalleled precision. By leveraging real-time data streams and behavioral insights, AI empowers brands to predict consumer intent, personalize engagement strategies, and optimize brand positioning. The chapter also addresses the ethical considerations, challenges of data governance, and the importance of human oversight in AI-driven marketing systems. Through illustrative case studies and emerging frameworks, this work highlights the transformative potential of AI in creating resilient, adaptive, and data-informed branding strategies suited for today's volatile and diverse consumer landscape.